1
0
Fork 0
mirror of https://github.com/warmcat/libwebsockets.git synced 2025-03-16 00:00:07 +01:00
libwebsockets/lib/parsers.c
Andy Green f94bef42c4 lws_header_table_reset: make caller responsibility to clear down ah rx buffer
There are two kinds of reaason to call lws_header_table_reset(), one is we are reallocating
a destroyed ah to another wsi, and the other is we are moving to the next pipelined header set
still on the same wsi, and we need a "weaker" reset that only clears down the state related
to the header parsing, not everything about the ah context including the ah rx buffer.

This patch moves the ah rxbuffer rxpos and rxlen resetting out of lws_header_table_reset() and to
be the responsibility of the caller.  Callers who are moving the ah to another wsi are
patched to deal with resetting rxpos and rxlen and lws_http_transaction_completed() who only
resets the ah when moving to the next pipelined headers, no longer wrongly clears the ah rxbuf.

https://github.com/warmcat/libwebsockets/issues/638
2016-09-29 10:31:06 +08:00

1521 lines
38 KiB
C

/*
* libwebsockets - small server side websockets and web server implementation
*
* Copyright (C) 2010-2013 Andy Green <andy@warmcat.com>
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation:
* version 2.1 of the License.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
* MA 02110-1301 USA
*/
#include "private-libwebsockets.h"
const unsigned char lextable[] = {
#include "lextable.h"
};
#define FAIL_CHAR 0x08
int LWS_WARN_UNUSED_RESULT
lextable_decode(int pos, char c)
{
if (c >= 'A' && c <= 'Z')
c += 'a' - 'A';
while (1) {
if (lextable[pos] & (1 << 7)) { /* 1-byte, fail on mismatch */
if ((lextable[pos] & 0x7f) != c)
return -1;
/* fall thru */
pos++;
if (lextable[pos] == FAIL_CHAR)
return -1;
return pos;
}
if (lextable[pos] == FAIL_CHAR)
return -1;
/* b7 = 0, end or 3-byte */
if (lextable[pos] < FAIL_CHAR) /* terminal marker */
return pos;
if (lextable[pos] == c) /* goto */
return pos + (lextable[pos + 1]) +
(lextable[pos + 2] << 8);
/* fall thru goto */
pos += 3;
/* continue */
}
}
// doesn't scrub the ah rxbuffer by default, parent must do if needed
void
lws_header_table_reset(struct lws *wsi, int autoservice)
{
struct allocated_headers *ah = wsi->u.hdr.ah;
struct lws_context_per_thread *pt;
struct lws_pollfd *pfd;
/* if we have the idea we're resetting 'our' ah, must be bound to one */
assert(ah);
/* ah also concurs with ownership */
assert(ah->wsi == wsi);
/* init the ah to reflect no headers or data have appeared yet */
memset(ah->frag_index, 0, sizeof(ah->frag_index));
ah->nfrag = 0;
ah->pos = 0;
/* since we will restart the ah, our new headers are not completed */
// wsi->hdr_parsing_completed = 0;
/*
* if we inherited pending rx (from socket adoption deferred
* processing), apply and free it.
*/
if (wsi->u.hdr.preamble_rx) {
memcpy(ah->rx, wsi->u.hdr.preamble_rx,
wsi->u.hdr.preamble_rx_len);
ah->rxlen = wsi->u.hdr.preamble_rx_len;
lws_free_set_NULL(wsi->u.hdr.preamble_rx);
if (autoservice) {
lwsl_notice("%s: calling service on readbuf ah\n", __func__);
pt = &wsi->context->pt[(int)wsi->tsi];
/* unlike a normal connect, we have the headers already
* (or the first part of them anyway)
*/
pfd = &pt->fds[wsi->position_in_fds_table];
pfd->revents |= LWS_POLLIN;
lwsl_err("%s: calling service\n", __func__);
lws_service_fd_tsi(wsi->context, pfd, wsi->tsi);
}
}
}
int LWS_WARN_UNUSED_RESULT
lws_header_table_attach(struct lws *wsi, int autoservice)
{
struct lws_context *context = wsi->context;
struct lws_context_per_thread *pt = &context->pt[(int)wsi->tsi];
struct lws_pollargs pa;
struct lws **pwsi;
int n;
lwsl_info("%s: wsi %p: ah %p (tsi %d, count = %d) in\n", __func__, (void *)wsi,
(void *)wsi->u.hdr.ah, wsi->tsi, pt->ah_count_in_use);
/* if we are already bound to one, just clear it down */
if (wsi->u.hdr.ah) {
lwsl_info("cleardown\n");
goto reset;
}
lws_pt_lock(pt);
pwsi = &pt->ah_wait_list;
while (*pwsi) {
if (*pwsi == wsi) {
/* if already waiting on list, if no new ah just ret */
if (pt->ah_count_in_use ==
context->max_http_header_pool) {
lwsl_notice("%s: no free ah to attach\n", __func__);
goto bail;
}
/* new ah.... remove ourselves from waiting list */
*pwsi = wsi->u.hdr.ah_wait_list; /* set our prev to our next */
wsi->u.hdr.ah_wait_list = NULL; /* no next any more */
pt->ah_wait_list_length--;
break;
}
pwsi = &(*pwsi)->u.hdr.ah_wait_list;
}
/*
* pool is all busy... add us to waiting list and return that we
* weren't able to deliver it right now
*/
if (pt->ah_count_in_use == context->max_http_header_pool) {
lwsl_notice("%s: adding %p to ah waiting list\n", __func__, wsi);
wsi->u.hdr.ah_wait_list = pt->ah_wait_list;
pt->ah_wait_list = wsi;
pt->ah_wait_list_length++;
/* we cannot accept input then */
_lws_change_pollfd(wsi, LWS_POLLIN, 0, &pa);
goto bail;
}
for (n = 0; n < context->max_http_header_pool; n++)
if (!pt->ah_pool[n].in_use)
break;
/* if the count of in use said something free... */
assert(n != context->max_http_header_pool);
wsi->u.hdr.ah = &pt->ah_pool[n];
wsi->u.hdr.ah->in_use = 1;
pt->ah_pool[n].wsi = wsi; /* mark our owner */
pt->ah_count_in_use++;
_lws_change_pollfd(wsi, 0, LWS_POLLIN, &pa);
lwsl_info("%s: wsi %p: ah %p: count %d (on exit)\n", __func__,
(void *)wsi, (void *)wsi->u.hdr.ah, pt->ah_count_in_use);
lws_pt_unlock(pt);
reset:
/* and reset the rx state */
wsi->u.hdr.ah->rxpos = 0;
wsi->u.hdr.ah->rxlen = 0;
lws_header_table_reset(wsi, autoservice);
time(&wsi->u.hdr.ah->assigned);
#ifndef LWS_NO_CLIENT
if (wsi->state == LWSS_CLIENT_UNCONNECTED)
if (!lws_client_connect_via_info2(wsi))
/* our client connect has failed, the wsi
* has been closed
*/
return -1;
#endif
return 0;
bail:
lws_pt_unlock(pt);
return 1;
}
int lws_header_table_detach(struct lws *wsi, int autoservice)
{
struct lws_context *context = wsi->context;
struct allocated_headers *ah = wsi->u.hdr.ah;
struct lws_context_per_thread *pt = &context->pt[(int)wsi->tsi];
struct lws_pollargs pa;
struct lws **pwsi;
time_t now;
lwsl_info("%s: wsi %p: ah %p (tsi=%d, count = %d)\n", __func__,
(void *)wsi, (void *)ah, wsi->tsi,
pt->ah_count_in_use);
if (wsi->u.hdr.preamble_rx)
lws_free_set_NULL(wsi->u.hdr.preamble_rx);
/* may not be detached while he still has unprocessed rx */
if (ah && ah->rxpos != ah->rxlen) {
lwsl_err("%s: %p: CANNOT DETACH rxpos:%d, rxlen:%d\n", __func__, wsi,
ah->rxpos, ah->rxlen);
assert(ah->rxpos == ah->rxlen);
return 0;
}
lws_pt_lock(pt);
pwsi = &pt->ah_wait_list;
if (!ah) { /* remove from wait list if none attached */
while (*pwsi) {
if (*pwsi == wsi) {
lwsl_info("%s: wsi %p, remv wait\n",
__func__, wsi);
*pwsi = wsi->u.hdr.ah_wait_list;
wsi->u.hdr.ah_wait_list = NULL;
pt->ah_wait_list_length--;
goto bail;
}
pwsi = &(*pwsi)->u.hdr.ah_wait_list;
}
/* no ah, not on list... no more business here */
goto bail;
}
/* we did have an ah attached */
time(&now);
if (ah->assigned && now - ah->assigned > 3) {
/*
* we're detaching the ah, but it was held an
* unreasonably long time
*/
lwsl_notice("%s: wsi %p: ah held %ds, "
"ah.rxpos %d, ah.rxlen %d, mode/state %d %d,"
"wsi->more_rx_waiting %d\n", __func__, wsi,
(int)(now - ah->assigned),
ah->rxpos, ah->rxlen, wsi->mode, wsi->state,
wsi->more_rx_waiting);
}
ah->assigned = 0;
/* if we think we're detaching one, there should be one in use */
assert(pt->ah_count_in_use > 0);
/* and this specific one should have been in use */
assert(ah->in_use);
wsi->u.hdr.ah = NULL;
ah->wsi = NULL; /* no owner */
/* oh there is nobody on the waiting list... leave it at that then */
if (!*pwsi) {
ah->in_use = 0;
pt->ah_count_in_use--;
goto bail;
}
/* somebody else on same tsi is waiting, give it to oldest guy */
lwsl_info("pt wait list %p\n", *pwsi);
while ((*pwsi)->u.hdr.ah_wait_list)
pwsi = &(*pwsi)->u.hdr.ah_wait_list;
wsi = *pwsi;
lwsl_info("last wsi in wait list %p\n", wsi);
wsi->u.hdr.ah = ah;
ah->wsi = wsi; /* new owner */
/* and reset the rx state */
ah->rxpos = 0;
ah->rxlen = 0;
lws_header_table_reset(wsi, autoservice);
time(&wsi->u.hdr.ah->assigned);
/* clients acquire the ah and then insert themselves in fds table... */
if (wsi->position_in_fds_table != -1) {
lwsl_info("%s: Enabling %p POLLIN\n", __func__, wsi);
/* he has been stuck waiting for an ah, but now his wait is over,
* let him progress
*/
_lws_change_pollfd(wsi, 0, LWS_POLLIN, &pa);
}
/* point prev guy to next guy in list instead */
*pwsi = wsi->u.hdr.ah_wait_list;
/* the guy who got one is out of the list */
wsi->u.hdr.ah_wait_list = NULL;
pt->ah_wait_list_length--;
#ifndef LWS_NO_CLIENT
if (wsi->state == LWSS_CLIENT_UNCONNECTED)
if (!lws_client_connect_via_info2(wsi)) {
/* our client connect has failed, the wsi
* has been closed
*/
lws_pt_unlock(pt);
return -1;
}
#endif
assert(!!pt->ah_wait_list_length == !!(int)(long)pt->ah_wait_list);
bail:
lwsl_info("%s: wsi %p: ah %p (tsi=%d, count = %d)\n", __func__,
(void *)wsi, (void *)ah, wsi->tsi,
pt->ah_count_in_use);
lws_pt_unlock(pt);
return 0;
}
LWS_VISIBLE int
lws_hdr_fragment_length(struct lws *wsi, enum lws_token_indexes h, int frag_idx)
{
int n;
n = wsi->u.hdr.ah->frag_index[h];
if (!n)
return 0;
do {
if (!frag_idx)
return wsi->u.hdr.ah->frags[n].len;
n = wsi->u.hdr.ah->frags[n].nfrag;
} while (frag_idx-- && n);
return 0;
}
LWS_VISIBLE int lws_hdr_total_length(struct lws *wsi, enum lws_token_indexes h)
{
int n;
int len = 0;
n = wsi->u.hdr.ah->frag_index[h];
if (!n)
return 0;
do {
len += wsi->u.hdr.ah->frags[n].len;
n = wsi->u.hdr.ah->frags[n].nfrag;
} while (n);
return len;
}
LWS_VISIBLE int lws_hdr_copy_fragment(struct lws *wsi, char *dst, int len,
enum lws_token_indexes h, int frag_idx)
{
int n = 0;
int f = wsi->u.hdr.ah->frag_index[h];
if (!f)
return -1;
while (n < frag_idx) {
f = wsi->u.hdr.ah->frags[f].nfrag;
if (!f)
return -1;
n++;
}
if (wsi->u.hdr.ah->frags[f].len >= len)
return -1;
memcpy(dst, wsi->u.hdr.ah->data + wsi->u.hdr.ah->frags[f].offset,
wsi->u.hdr.ah->frags[f].len);
dst[wsi->u.hdr.ah->frags[f].len] = '\0';
return wsi->u.hdr.ah->frags[f].len;
}
LWS_VISIBLE int lws_hdr_copy(struct lws *wsi, char *dst, int len,
enum lws_token_indexes h)
{
int toklen = lws_hdr_total_length(wsi, h);
int n;
if (toklen >= len)
return -1;
n = wsi->u.hdr.ah->frag_index[h];
if (!n)
return 0;
do {
strcpy(dst, &wsi->u.hdr.ah->data[wsi->u.hdr.ah->frags[n].offset]);
dst += wsi->u.hdr.ah->frags[n].len;
n = wsi->u.hdr.ah->frags[n].nfrag;
} while (n);
return toklen;
}
char *lws_hdr_simple_ptr(struct lws *wsi, enum lws_token_indexes h)
{
int n;
n = wsi->u.hdr.ah->frag_index[h];
if (!n)
return NULL;
return wsi->u.hdr.ah->data + wsi->u.hdr.ah->frags[n].offset;
}
int LWS_WARN_UNUSED_RESULT
lws_pos_in_bounds(struct lws *wsi)
{
if (wsi->u.hdr.ah->pos < (unsigned int)wsi->context->max_http_header_data)
return 0;
if (wsi->u.hdr.ah->pos == wsi->context->max_http_header_data) {
lwsl_err("Ran out of header data space\n");
return 1;
}
/*
* with these tests everywhere, it should never be able to exceed
* the limit, only meet the limit
*/
lwsl_err("%s: pos %d, limit %d\n", __func__, wsi->u.hdr.ah->pos,
wsi->context->max_http_header_data);
assert(0);
return 1;
}
int LWS_WARN_UNUSED_RESULT
lws_hdr_simple_create(struct lws *wsi, enum lws_token_indexes h, const char *s)
{
wsi->u.hdr.ah->nfrag++;
if (wsi->u.hdr.ah->nfrag == ARRAY_SIZE(wsi->u.hdr.ah->frags)) {
lwsl_warn("More hdr frags than we can deal with, dropping\n");
return -1;
}
wsi->u.hdr.ah->frag_index[h] = wsi->u.hdr.ah->nfrag;
wsi->u.hdr.ah->frags[wsi->u.hdr.ah->nfrag].offset = wsi->u.hdr.ah->pos;
wsi->u.hdr.ah->frags[wsi->u.hdr.ah->nfrag].len = 0;
wsi->u.hdr.ah->frags[wsi->u.hdr.ah->nfrag].nfrag = 0;
do {
if (lws_pos_in_bounds(wsi))
return -1;
wsi->u.hdr.ah->data[wsi->u.hdr.ah->pos++] = *s;
if (*s)
wsi->u.hdr.ah->frags[wsi->u.hdr.ah->nfrag].len++;
} while (*s++);
return 0;
}
signed char char_to_hex(const char c)
{
if (c >= '0' && c <= '9')
return c - '0';
if (c >= 'a' && c <= 'f')
return c - 'a' + 10;
if (c >= 'A' && c <= 'F')
return c - 'A' + 10;
return -1;
}
static int LWS_WARN_UNUSED_RESULT
issue_char(struct lws *wsi, unsigned char c)
{
unsigned short frag_len;
if (lws_pos_in_bounds(wsi))
return -1;
frag_len = wsi->u.hdr.ah->frags[wsi->u.hdr.ah->nfrag].len;
/*
* If we haven't hit the token limit, just copy the character into
* the header
*/
if (frag_len < wsi->u.hdr.current_token_limit) {
wsi->u.hdr.ah->data[wsi->u.hdr.ah->pos++] = c;
if (c)
wsi->u.hdr.ah->frags[wsi->u.hdr.ah->nfrag].len++;
return 0;
}
/* Insert a null character when we *hit* the limit: */
if (frag_len == wsi->u.hdr.current_token_limit) {
if (lws_pos_in_bounds(wsi))
return -1;
wsi->u.hdr.ah->data[wsi->u.hdr.ah->pos++] = '\0';
lwsl_warn("header %i exceeds limit %d\n",
wsi->u.hdr.parser_state,
wsi->u.hdr.current_token_limit);
}
return 1;
}
int LWS_WARN_UNUSED_RESULT
lws_parse(struct lws *wsi, unsigned char c)
{
static const unsigned char methods[] = {
WSI_TOKEN_GET_URI,
WSI_TOKEN_POST_URI,
WSI_TOKEN_OPTIONS_URI,
WSI_TOKEN_PUT_URI,
WSI_TOKEN_PATCH_URI,
WSI_TOKEN_DELETE_URI,
};
struct allocated_headers *ah = wsi->u.hdr.ah;
struct lws_context *context = wsi->context;
unsigned int n, m, enc = 0;
assert(wsi->u.hdr.ah);
switch (wsi->u.hdr.parser_state) {
default:
lwsl_parser("WSI_TOK_(%d) '%c'\n", wsi->u.hdr.parser_state, c);
/* collect into malloc'd buffers */
/* optional initial space swallow */
if (!ah->frags[ah->frag_index[wsi->u.hdr.parser_state]].len &&
c == ' ')
break;
for (m = 0; m < ARRAY_SIZE(methods); m++)
if (wsi->u.hdr.parser_state == methods[m])
break;
if (m == ARRAY_SIZE(methods))
/* it was not any of the methods */
goto check_eol;
/* special URI processing... end at space */
if (c == ' ') {
/* enforce starting with / */
if (!ah->frags[ah->nfrag].len)
if (issue_char(wsi, '/') < 0)
return -1;
if (wsi->u.hdr.ups == URIPS_SEEN_SLASH_DOT_DOT) {
/*
* back up one dir level if possible
* safe against header fragmentation because
* the method URI can only be in 1 fragment
*/
if (ah->frags[ah->nfrag].len > 2) {
ah->pos--;
ah->frags[ah->nfrag].len--;
do {
ah->pos--;
ah->frags[ah->nfrag].len--;
} while (ah->frags[ah->nfrag].len > 1 &&
ah->data[ah->pos] != '/');
}
}
/* begin parsing HTTP version: */
if (issue_char(wsi, '\0') < 0)
return -1;
wsi->u.hdr.parser_state = WSI_TOKEN_HTTP;
goto start_fragment;
}
/*
* PRIORITY 1
* special URI processing... convert %xx
*/
switch (wsi->u.hdr.ues) {
case URIES_IDLE:
if (c == '%') {
wsi->u.hdr.ues = URIES_SEEN_PERCENT;
goto swallow;
}
break;
case URIES_SEEN_PERCENT:
if (char_to_hex(c) < 0)
/* illegal post-% char */
goto forbid;
wsi->u.hdr.esc_stash = c;
wsi->u.hdr.ues = URIES_SEEN_PERCENT_H1;
goto swallow;
case URIES_SEEN_PERCENT_H1:
if (char_to_hex(c) < 0)
/* illegal post-% char */
goto forbid;
c = (char_to_hex(wsi->u.hdr.esc_stash) << 4) |
char_to_hex(c);
enc = 1;
wsi->u.hdr.ues = URIES_IDLE;
break;
}
/*
* PRIORITY 2
* special URI processing...
* convert /.. or /... or /../ etc to /
* convert /./ to /
* convert // or /// etc to /
* leave /.dir or whatever alone
*/
switch (wsi->u.hdr.ups) {
case URIPS_IDLE:
if (!c)
return -1;
/* genuine delimiter */
if ((c == '&' || c == ';') && !enc) {
if (issue_char(wsi, c) < 0)
return -1;
/* swallow the terminator */
ah->frags[ah->nfrag].len--;
/* link to next fragment */
ah->frags[ah->nfrag].nfrag = ah->nfrag + 1;
ah->nfrag++;
if (ah->nfrag >= ARRAY_SIZE(ah->frags))
goto excessive;
/* start next fragment after the & */
wsi->u.hdr.post_literal_equal = 0;
ah->frags[ah->nfrag].offset = ah->pos;
ah->frags[ah->nfrag].len = 0;
ah->frags[ah->nfrag].nfrag = 0;
goto swallow;
}
/* uriencoded = in the name part, disallow */
if (c == '=' && enc &&
ah->frag_index[WSI_TOKEN_HTTP_URI_ARGS] &&
!wsi->u.hdr.post_literal_equal)
c = '_';
/* after the real =, we don't care how many = */
if (c == '=' && !enc)
wsi->u.hdr.post_literal_equal = 1;
/* + to space */
if (c == '+' && !enc)
c = ' ';
/* issue the first / always */
if (c == '/' && !ah->frag_index[WSI_TOKEN_HTTP_URI_ARGS])
wsi->u.hdr.ups = URIPS_SEEN_SLASH;
break;
case URIPS_SEEN_SLASH:
/* swallow subsequent slashes */
if (c == '/')
goto swallow;
/* track and swallow the first . after / */
if (c == '.') {
wsi->u.hdr.ups = URIPS_SEEN_SLASH_DOT;
goto swallow;
}
wsi->u.hdr.ups = URIPS_IDLE;
break;
case URIPS_SEEN_SLASH_DOT:
/* swallow second . */
if (c == '.') {
wsi->u.hdr.ups = URIPS_SEEN_SLASH_DOT_DOT;
goto swallow;
}
/* change /./ to / */
if (c == '/') {
wsi->u.hdr.ups = URIPS_SEEN_SLASH;
goto swallow;
}
/* it was like /.dir ... regurgitate the . */
wsi->u.hdr.ups = URIPS_IDLE;
if (issue_char(wsi, '.') < 0)
return -1;
break;
case URIPS_SEEN_SLASH_DOT_DOT:
/* /../ or /..[End of URI] --> backup to last / */
if (c == '/' || c == '?') {
/*
* back up one dir level if possible
* safe against header fragmentation because
* the method URI can only be in 1 fragment
*/
if (ah->frags[ah->nfrag].len > 2) {
ah->pos--;
ah->frags[ah->nfrag].len--;
do {
ah->pos--;
ah->frags[ah->nfrag].len--;
} while (ah->frags[ah->nfrag].len > 1 &&
ah->data[ah->pos] != '/');
}
wsi->u.hdr.ups = URIPS_SEEN_SLASH;
if (ah->frags[ah->nfrag].len > 1)
break;
goto swallow;
}
/* /..[^/] ... regurgitate and allow */
if (issue_char(wsi, '.') < 0)
return -1;
if (issue_char(wsi, '.') < 0)
return -1;
wsi->u.hdr.ups = URIPS_IDLE;
break;
}
if (c == '?' && !enc &&
!ah->frag_index[WSI_TOKEN_HTTP_URI_ARGS]) { /* start of URI arguments */
if (wsi->u.hdr.ues != URIES_IDLE)
goto forbid;
/* seal off uri header */
if (issue_char(wsi, '\0') < 0)
return -1;
/* move to using WSI_TOKEN_HTTP_URI_ARGS */
ah->nfrag++;
if (ah->nfrag >= ARRAY_SIZE(ah->frags))
goto excessive;
ah->frags[ah->nfrag].offset = ah->pos;
ah->frags[ah->nfrag].len = 0;
ah->frags[ah->nfrag].nfrag = 0;
wsi->u.hdr.post_literal_equal = 0;
ah->frag_index[WSI_TOKEN_HTTP_URI_ARGS] = ah->nfrag;
wsi->u.hdr.ups = URIPS_IDLE;
goto swallow;
}
check_eol:
/* bail at EOL */
if (wsi->u.hdr.parser_state != WSI_TOKEN_CHALLENGE &&
c == '\x0d') {
if (wsi->u.hdr.ues != URIES_IDLE)
goto forbid;
c = '\0';
wsi->u.hdr.parser_state = WSI_TOKEN_SKIPPING_SAW_CR;
lwsl_parser("*\n");
}
n = issue_char(wsi, c);
if ((int)n < 0)
return -1;
if (n > 0)
wsi->u.hdr.parser_state = WSI_TOKEN_SKIPPING;
swallow:
/* per-protocol end of headers management */
if (wsi->u.hdr.parser_state == WSI_TOKEN_CHALLENGE)
goto set_parsing_complete;
break;
/* collecting and checking a name part */
case WSI_TOKEN_NAME_PART:
lwsl_parser("WSI_TOKEN_NAME_PART '%c' (mode=%d)\n", c, wsi->mode);
wsi->u.hdr.lextable_pos =
lextable_decode(wsi->u.hdr.lextable_pos, c);
/*
* Server needs to look out for unknown methods...
*/
if (wsi->u.hdr.lextable_pos < 0 &&
wsi->mode == LWSCM_HTTP_SERVING) {
/* this is not a header we know about */
for (m = 0; m < ARRAY_SIZE(methods); m++)
if (ah->frag_index[methods[m]]) {
/*
* already had the method, no idea what
* this crap from the client is, ignore
*/
wsi->u.hdr.parser_state = WSI_TOKEN_SKIPPING;
break;
}
/*
* hm it's an unknown http method from a client in fact,
* treat as dangerous
*/
if (m == ARRAY_SIZE(methods)) {
lwsl_info("Unknown method - dropping\n");
goto forbid;
}
break;
}
/*
* ...otherwise for a client, let him ignore unknown headers
* coming from the server
*/
if (wsi->u.hdr.lextable_pos < 0) {
wsi->u.hdr.parser_state = WSI_TOKEN_SKIPPING;
break;
}
if (lextable[wsi->u.hdr.lextable_pos] < FAIL_CHAR) {
/* terminal state */
n = ((unsigned int)lextable[wsi->u.hdr.lextable_pos] << 8) |
lextable[wsi->u.hdr.lextable_pos + 1];
lwsl_parser("known hdr %d\n", n);
for (m = 0; m < ARRAY_SIZE(methods); m++)
if (n == methods[m] &&
ah->frag_index[methods[m]]) {
lwsl_warn("Duplicated method\n");
return -1;
}
/*
* WSORIGIN is protocol equiv to ORIGIN,
* JWebSocket likes to send it, map to ORIGIN
*/
if (n == WSI_TOKEN_SWORIGIN)
n = WSI_TOKEN_ORIGIN;
wsi->u.hdr.parser_state = (enum lws_token_indexes)
(WSI_TOKEN_GET_URI + n);
if (context->token_limits)
wsi->u.hdr.current_token_limit =
context->token_limits->token_limit[
wsi->u.hdr.parser_state];
else
wsi->u.hdr.current_token_limit =
wsi->context->max_http_header_data;
if (wsi->u.hdr.parser_state == WSI_TOKEN_CHALLENGE)
goto set_parsing_complete;
goto start_fragment;
}
break;
start_fragment:
ah->nfrag++;
excessive:
if (ah->nfrag == ARRAY_SIZE(ah->frags)) {
lwsl_warn("More hdr frags than we can deal with\n");
return -1;
}
ah->frags[ah->nfrag].offset = ah->pos;
ah->frags[ah->nfrag].len = 0;
ah->frags[ah->nfrag].nfrag = 0;
n = ah->frag_index[wsi->u.hdr.parser_state];
if (!n) { /* first fragment */
ah->frag_index[wsi->u.hdr.parser_state] = ah->nfrag;
break;
}
/* continuation */
while (ah->frags[n].nfrag)
n = ah->frags[n].nfrag;
ah->frags[n].nfrag = ah->nfrag;
if (issue_char(wsi, ' ') < 0)
return -1;
break;
/* skipping arg part of a name we didn't recognize */
case WSI_TOKEN_SKIPPING:
lwsl_parser("WSI_TOKEN_SKIPPING '%c'\n", c);
if (c == '\x0d')
wsi->u.hdr.parser_state = WSI_TOKEN_SKIPPING_SAW_CR;
break;
case WSI_TOKEN_SKIPPING_SAW_CR:
lwsl_parser("WSI_TOKEN_SKIPPING_SAW_CR '%c'\n", c);
if (wsi->u.hdr.ues != URIES_IDLE)
goto forbid;
if (c == '\x0a') {
wsi->u.hdr.parser_state = WSI_TOKEN_NAME_PART;
wsi->u.hdr.lextable_pos = 0;
} else
wsi->u.hdr.parser_state = WSI_TOKEN_SKIPPING;
break;
/* we're done, ignore anything else */
case WSI_PARSING_COMPLETE:
lwsl_parser("WSI_PARSING_COMPLETE '%c'\n", c);
break;
}
return 0;
set_parsing_complete:
if (wsi->u.hdr.ues != URIES_IDLE)
goto forbid;
if (lws_hdr_total_length(wsi, WSI_TOKEN_UPGRADE)) {
if (lws_hdr_total_length(wsi, WSI_TOKEN_VERSION))
wsi->ietf_spec_revision =
atoi(lws_hdr_simple_ptr(wsi, WSI_TOKEN_VERSION));
lwsl_parser("v%02d hdrs completed\n", wsi->ietf_spec_revision);
}
wsi->u.hdr.parser_state = WSI_PARSING_COMPLETE;
wsi->hdr_parsing_completed = 1;
return 0;
forbid:
lwsl_notice(" forbidding on uri sanitation\n");
lws_return_http_status(wsi, HTTP_STATUS_FORBIDDEN, NULL);
return -1;
}
LWS_VISIBLE int lws_frame_is_binary(struct lws *wsi)
{
return wsi->u.ws.frame_is_binary;
}
int
lws_rx_sm(struct lws *wsi, unsigned char c)
{
struct lws_context_per_thread *pt = &wsi->context->pt[(int)wsi->tsi];
int callback_action = LWS_CALLBACK_RECEIVE;
int ret = 0, n, rx_draining_ext = 0;
struct lws_tokens eff_buf;
if (wsi->socket_is_permanently_unusable)
return -1;
switch (wsi->lws_rx_parse_state) {
case LWS_RXPS_NEW:
if (wsi->u.ws.rx_draining_ext) {
struct lws **w = &pt->rx_draining_ext_list;
eff_buf.token = NULL;
eff_buf.token_len = 0;
wsi->u.ws.rx_draining_ext = 0;
/* remove us from context draining ext list */
while (*w) {
if (*w == wsi) {
*w = wsi->u.ws.rx_draining_ext_list;
break;
}
w = &((*w)->u.ws.rx_draining_ext_list);
}
wsi->u.ws.rx_draining_ext_list = NULL;
rx_draining_ext = 1;
lwsl_err("%s: doing draining flow\n", __func__);
goto drain_extension;
}
switch (wsi->ietf_spec_revision) {
case 13:
/*
* no prepended frame key any more
*/
wsi->u.ws.all_zero_nonce = 1;
goto handle_first;
default:
lwsl_warn("lws_rx_sm: unknown spec version %d\n",
wsi->ietf_spec_revision);
break;
}
break;
case LWS_RXPS_04_mask_1:
wsi->u.ws.mask[1] = c;
if (c)
wsi->u.ws.all_zero_nonce = 0;
wsi->lws_rx_parse_state = LWS_RXPS_04_mask_2;
break;
case LWS_RXPS_04_mask_2:
wsi->u.ws.mask[2] = c;
if (c)
wsi->u.ws.all_zero_nonce = 0;
wsi->lws_rx_parse_state = LWS_RXPS_04_mask_3;
break;
case LWS_RXPS_04_mask_3:
wsi->u.ws.mask[3] = c;
if (c)
wsi->u.ws.all_zero_nonce = 0;
/*
* start from the zero'th byte in the XOR key buffer since
* this is the start of a frame with a new key
*/
wsi->u.ws.mask_idx = 0;
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_1;
break;
/*
* 04 logical framing from the spec (all this is masked when incoming
* and has to be unmasked)
*
* We ignore the possibility of extension data because we don't
* negotiate any extensions at the moment.
*
* 0 1 2 3
* 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
* +-+-+-+-+-------+-+-------------+-------------------------------+
* |F|R|R|R| opcode|R| Payload len | Extended payload length |
* |I|S|S|S| (4) |S| (7) | (16/63) |
* |N|V|V|V| |V| | (if payload len==126/127) |
* | |1|2|3| |4| | |
* +-+-+-+-+-------+-+-------------+ - - - - - - - - - - - - - - - +
* | Extended payload length continued, if payload len == 127 |
* + - - - - - - - - - - - - - - - +-------------------------------+
* | | Extension data |
* +-------------------------------+ - - - - - - - - - - - - - - - +
* : :
* +---------------------------------------------------------------+
* : Application data :
* +---------------------------------------------------------------+
*
* We pass payload through to userland as soon as we get it, ignoring
* FIN. It's up to userland to buffer it up if it wants to see a
* whole unfragmented block of the original size (which may be up to
* 2^63 long!)
*/
case LWS_RXPS_04_FRAME_HDR_1:
handle_first:
wsi->u.ws.opcode = c & 0xf;
wsi->u.ws.rsv = c & 0x70;
wsi->u.ws.final = !!((c >> 7) & 1);
switch (wsi->u.ws.opcode) {
case LWSWSOPC_TEXT_FRAME:
case LWSWSOPC_BINARY_FRAME:
wsi->u.ws.rsv_first_msg = (c & 0x70);
wsi->u.ws.frame_is_binary =
wsi->u.ws.opcode == LWSWSOPC_BINARY_FRAME;
break;
case 3:
case 4:
case 5:
case 6:
case 7:
case 0xb:
case 0xc:
case 0xd:
case 0xe:
case 0xf:
lwsl_info("illegal opcode\n");
return -1;
}
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN;
break;
case LWS_RXPS_04_FRAME_HDR_LEN:
wsi->u.ws.this_frame_masked = !!(c & 0x80);
switch (c & 0x7f) {
case 126:
/* control frames are not allowed to have big lengths */
if (wsi->u.ws.opcode & 8)
goto illegal_ctl_length;
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN16_2;
break;
case 127:
/* control frames are not allowed to have big lengths */
if (wsi->u.ws.opcode & 8)
goto illegal_ctl_length;
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_8;
break;
default:
wsi->u.ws.rx_packet_length = c & 0x7f;
if (wsi->u.ws.this_frame_masked)
wsi->lws_rx_parse_state =
LWS_RXPS_07_COLLECT_FRAME_KEY_1;
else
if (wsi->u.ws.rx_packet_length)
wsi->lws_rx_parse_state =
LWS_RXPS_PAYLOAD_UNTIL_LENGTH_EXHAUSTED;
else {
wsi->lws_rx_parse_state = LWS_RXPS_NEW;
goto spill;
}
break;
}
break;
case LWS_RXPS_04_FRAME_HDR_LEN16_2:
wsi->u.ws.rx_packet_length = c << 8;
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN16_1;
break;
case LWS_RXPS_04_FRAME_HDR_LEN16_1:
wsi->u.ws.rx_packet_length |= c;
if (wsi->u.ws.this_frame_masked)
wsi->lws_rx_parse_state =
LWS_RXPS_07_COLLECT_FRAME_KEY_1;
else
wsi->lws_rx_parse_state =
LWS_RXPS_PAYLOAD_UNTIL_LENGTH_EXHAUSTED;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_8:
if (c & 0x80) {
lwsl_warn("b63 of length must be zero\n");
/* kill the connection */
return -1;
}
#if defined __LP64__
wsi->u.ws.rx_packet_length = ((size_t)c) << 56;
#else
wsi->u.ws.rx_packet_length = 0;
#endif
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_7;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_7:
#if defined __LP64__
wsi->u.ws.rx_packet_length |= ((size_t)c) << 48;
#endif
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_6;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_6:
#if defined __LP64__
wsi->u.ws.rx_packet_length |= ((size_t)c) << 40;
#endif
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_5;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_5:
#if defined __LP64__
wsi->u.ws.rx_packet_length |= ((size_t)c) << 32;
#endif
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_4;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_4:
wsi->u.ws.rx_packet_length |= ((size_t)c) << 24;
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_3;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_3:
wsi->u.ws.rx_packet_length |= ((size_t)c) << 16;
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_2;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_2:
wsi->u.ws.rx_packet_length |= ((size_t)c) << 8;
wsi->lws_rx_parse_state = LWS_RXPS_04_FRAME_HDR_LEN64_1;
break;
case LWS_RXPS_04_FRAME_HDR_LEN64_1:
wsi->u.ws.rx_packet_length |= ((size_t)c);
if (wsi->u.ws.this_frame_masked)
wsi->lws_rx_parse_state =
LWS_RXPS_07_COLLECT_FRAME_KEY_1;
else
wsi->lws_rx_parse_state =
LWS_RXPS_PAYLOAD_UNTIL_LENGTH_EXHAUSTED;
break;
case LWS_RXPS_07_COLLECT_FRAME_KEY_1:
wsi->u.ws.mask[0] = c;
if (c)
wsi->u.ws.all_zero_nonce = 0;
wsi->lws_rx_parse_state = LWS_RXPS_07_COLLECT_FRAME_KEY_2;
break;
case LWS_RXPS_07_COLLECT_FRAME_KEY_2:
wsi->u.ws.mask[1] = c;
if (c)
wsi->u.ws.all_zero_nonce = 0;
wsi->lws_rx_parse_state = LWS_RXPS_07_COLLECT_FRAME_KEY_3;
break;
case LWS_RXPS_07_COLLECT_FRAME_KEY_3:
wsi->u.ws.mask[2] = c;
if (c)
wsi->u.ws.all_zero_nonce = 0;
wsi->lws_rx_parse_state = LWS_RXPS_07_COLLECT_FRAME_KEY_4;
break;
case LWS_RXPS_07_COLLECT_FRAME_KEY_4:
wsi->u.ws.mask[3] = c;
if (c)
wsi->u.ws.all_zero_nonce = 0;
wsi->lws_rx_parse_state =
LWS_RXPS_PAYLOAD_UNTIL_LENGTH_EXHAUSTED;
wsi->u.ws.mask_idx = 0;
if (wsi->u.ws.rx_packet_length == 0) {
wsi->lws_rx_parse_state = LWS_RXPS_NEW;
goto spill;
}
break;
case LWS_RXPS_PAYLOAD_UNTIL_LENGTH_EXHAUSTED:
assert(wsi->u.ws.rx_ubuf);
if (wsi->u.ws.rx_ubuf_head + LWS_PRE >=
wsi->u.ws.rx_ubuf_alloc) {
lwsl_err("Attempted overflow \n");
return -1;
}
if (wsi->u.ws.all_zero_nonce)
wsi->u.ws.rx_ubuf[LWS_PRE +
(wsi->u.ws.rx_ubuf_head++)] = c;
else
wsi->u.ws.rx_ubuf[LWS_PRE +
(wsi->u.ws.rx_ubuf_head++)] =
c ^ wsi->u.ws.mask[
(wsi->u.ws.mask_idx++) & 3];
if (--wsi->u.ws.rx_packet_length == 0) {
/* spill because we have the whole frame */
wsi->lws_rx_parse_state = LWS_RXPS_NEW;
goto spill;
}
/*
* if there's no protocol max frame size given, we are
* supposed to default to context->pt_serv_buf_size
*/
if (!wsi->protocol->rx_buffer_size &&
wsi->u.ws.rx_ubuf_head != wsi->context->pt_serv_buf_size)
break;
else
if (wsi->protocol->rx_buffer_size &&
wsi->u.ws.rx_ubuf_head !=
wsi->protocol->rx_buffer_size)
break;
/* spill because we filled our rx buffer */
spill:
/*
* is this frame a control packet we should take care of at this
* layer? If so service it and hide it from the user callback
*/
lwsl_parser("spill on %s\n", wsi->protocol->name);
switch (wsi->u.ws.opcode) {
case LWSWSOPC_CLOSE:
/* is this an acknowledgement of our close? */
if (wsi->state == LWSS_AWAITING_CLOSE_ACK) {
/*
* fine he has told us he is closing too, let's
* finish our close
*/
lwsl_parser("seen client close ack\n");
return -1;
}
if (wsi->state == LWSS_RETURNED_CLOSE_ALREADY)
/* if he sends us 2 CLOSE, kill him */
return -1;
if (lws_partial_buffered(wsi)) {
/*
* if we're in the middle of something,
* we can't do a normal close response and
* have to just close our end.
*/
wsi->socket_is_permanently_unusable = 1;
lwsl_parser("Closing on peer close due to Pending tx\n");
return -1;
}
if (user_callback_handle_rxflow(
wsi->protocol->callback, wsi,
LWS_CALLBACK_WS_PEER_INITIATED_CLOSE,
wsi->user_space,
&wsi->u.ws.rx_ubuf[LWS_PRE],
wsi->u.ws.rx_ubuf_head))
return -1;
lwsl_parser("server sees client close packet\n");
wsi->state = LWSS_RETURNED_CLOSE_ALREADY;
/* deal with the close packet contents as a PONG */
wsi->u.ws.payload_is_close = 1;
goto process_as_ping;
case LWSWSOPC_PING:
lwsl_info("received %d byte ping, sending pong\n",
wsi->u.ws.rx_ubuf_head);
if (wsi->u.ws.ping_pending_flag) {
/*
* there is already a pending ping payload
* we should just log and drop
*/
lwsl_parser("DROP PING since one pending\n");
goto ping_drop;
}
process_as_ping:
/* control packets can only be < 128 bytes long */
if (wsi->u.ws.rx_ubuf_head > 128 - 3) {
lwsl_parser("DROP PING payload too large\n");
goto ping_drop;
}
/* stash the pong payload */
memcpy(wsi->u.ws.ping_payload_buf + LWS_PRE,
&wsi->u.ws.rx_ubuf[LWS_PRE],
wsi->u.ws.rx_ubuf_head);
wsi->u.ws.ping_payload_len = wsi->u.ws.rx_ubuf_head;
wsi->u.ws.ping_pending_flag = 1;
/* get it sent as soon as possible */
lws_callback_on_writable(wsi);
ping_drop:
wsi->u.ws.rx_ubuf_head = 0;
return 0;
case LWSWSOPC_PONG:
lwsl_info("received pong\n");
lwsl_hexdump(&wsi->u.ws.rx_ubuf[LWS_PRE],
wsi->u.ws.rx_ubuf_head);
if (wsi->pending_timeout == PENDING_TIMEOUT_WS_PONG_CHECK_GET_PONG) {
lwsl_info("received expected PONG on wsi %p\n", wsi);
lws_set_timeout(wsi, NO_PENDING_TIMEOUT, 0);
}
/* issue it */
callback_action = LWS_CALLBACK_RECEIVE_PONG;
break;
case LWSWSOPC_TEXT_FRAME:
case LWSWSOPC_BINARY_FRAME:
case LWSWSOPC_CONTINUATION:
break;
default:
lwsl_parser("passing opc %x up to exts\n",
wsi->u.ws.opcode);
/*
* It's something special we can't understand here.
* Pass the payload up to the extension's parsing
* state machine.
*/
eff_buf.token = &wsi->u.ws.rx_ubuf[LWS_PRE];
eff_buf.token_len = wsi->u.ws.rx_ubuf_head;
if (lws_ext_cb_active(wsi, LWS_EXT_CB_EXTENDED_PAYLOAD_RX,
&eff_buf, 0) <= 0)
/* not handle or fail */
lwsl_ext("ext opc opcode 0x%x unknown\n",
wsi->u.ws.opcode);
wsi->u.ws.rx_ubuf_head = 0;
return 0;
}
/*
* No it's real payload, pass it up to the user callback.
* It's nicely buffered with the pre-padding taken care of
* so it can be sent straight out again using lws_write
*/
eff_buf.token = &wsi->u.ws.rx_ubuf[LWS_PRE];
eff_buf.token_len = wsi->u.ws.rx_ubuf_head;
drain_extension:
lwsl_ext("%s: passing %d to ext\n", __func__, eff_buf.token_len);
if (wsi->state == LWSS_RETURNED_CLOSE_ALREADY ||
wsi->state == LWSS_AWAITING_CLOSE_ACK)
goto already_done;
n = lws_ext_cb_active(wsi, LWS_EXT_CB_PAYLOAD_RX, &eff_buf, 0);
if (n < 0) {
/*
* we may rely on this to get RX, just drop connection
*/
wsi->socket_is_permanently_unusable = 1;
return -1;
}
if (rx_draining_ext && eff_buf.token_len == 0)
goto already_done;
if (n && eff_buf.token_len) {
/* extension had more... main loop will come back */
wsi->u.ws.rx_draining_ext = 1;
wsi->u.ws.rx_draining_ext_list = pt->rx_draining_ext_list;
pt->rx_draining_ext_list = wsi;
}
if (eff_buf.token_len > 0 ||
callback_action == LWS_CALLBACK_RECEIVE_PONG) {
eff_buf.token[eff_buf.token_len] = '\0';
if (wsi->protocol->callback) {
if (callback_action == LWS_CALLBACK_RECEIVE_PONG)
lwsl_info("Doing pong callback\n");
ret = user_callback_handle_rxflow(
wsi->protocol->callback,
wsi,
(enum lws_callback_reasons)callback_action,
wsi->user_space,
eff_buf.token,
eff_buf.token_len);
}
else
lwsl_err("No callback on payload spill!\n");
}
already_done:
wsi->u.ws.rx_ubuf_head = 0;
break;
}
return ret;
illegal_ctl_length:
lwsl_warn("Control frame with xtended length is illegal\n");
/* kill the connection */
return -1;
}
LWS_VISIBLE size_t
lws_remaining_packet_payload(struct lws *wsi)
{
return wsi->u.ws.rx_packet_length;
}
/* Once we reach LWS_RXPS_PAYLOAD_UNTIL_LENGTH_EXHAUSTED, we know how much
* to expect in that state and can deal with it in bulk more efficiently.
*/
int
lws_payload_until_length_exhausted(struct lws *wsi, unsigned char **buf,
size_t *len)
{
unsigned char *buffer = *buf, mask[4];
int buffer_size, n;
unsigned int avail;
char *rx_ubuf;
if (wsi->protocol->rx_buffer_size)
buffer_size = wsi->protocol->rx_buffer_size;
else
buffer_size = wsi->context->pt_serv_buf_size;
avail = buffer_size - wsi->u.ws.rx_ubuf_head;
/* do not consume more than we should */
if (avail > wsi->u.ws.rx_packet_length)
avail = wsi->u.ws.rx_packet_length;
/* do not consume more than what is in the buffer */
if (avail > *len)
avail = *len;
/* we want to leave 1 byte for the parser to handle properly */
if (avail <= 1)
return 0;
avail--;
rx_ubuf = wsi->u.ws.rx_ubuf + LWS_PRE + wsi->u.ws.rx_ubuf_head;
if (wsi->u.ws.all_zero_nonce)
memcpy(rx_ubuf, buffer, avail);
else {
for (n = 0; n < 4; n++)
mask[n] = wsi->u.ws.mask[(wsi->u.ws.mask_idx + n) & 3];
/* deal with 4-byte chunks using unwrapped loop */
n = avail >> 2;
while (n--) {
*(rx_ubuf++) = *(buffer++) ^ mask[0];
*(rx_ubuf++) = *(buffer++) ^ mask[1];
*(rx_ubuf++) = *(buffer++) ^ mask[2];
*(rx_ubuf++) = *(buffer++) ^ mask[3];
}
/* and the remaining bytes bytewise */
for (n = 0; n < (int)(avail & 3); n++)
*(rx_ubuf++) = *(buffer++) ^ mask[n];
wsi->u.ws.mask_idx = (wsi->u.ws.mask_idx + avail) & 3;
}
(*buf) += avail;
wsi->u.ws.rx_ubuf_head += avail;
wsi->u.ws.rx_packet_length -= avail;
*len -= avail;
return avail;
}